[Qemu-devel] Re: [PATCH 10/14] Zero json struct with memset() instea of = {} to keep compiler happy.

[Anthony Liguori <anthony@codemonkey.ws>]

On 08/30/2010 10:43 AM, Jes Sorensen wrote:
> On 08/30/10 17:39, Anthony Liguori wrote:
>    
>> On 08/30/2010 10:35 AM, Jes.Sorensen@redhat.com wrote:
>>      
>>> From: Jes Sorensen<Jes.Sorensen@redhat.com>
>>>
>>> This keeps the compiler happy when building with -Wextra while
>>> effectively generating the same code.
>>>
>>> Signed-off-by: Jes Sorensen<Jes.Sorensen@redhat.com>
>>>
>>>        
>> What's GCC's compliant?
>>      
> cc1: warnings being treated as errors
> qjson.c: In function 'qobject_from_jsonv':
> qjson.c:39: error: missing initializer
> qjson.c:39: error: (near initialization for 'state.parser')
> make: *** [qjson.o] Error 1
>
> We have a lot of these where we try to init a struct element {}.
>
> Yes it's technically legal. However it's painful when you try to apply
> more aggressive warning flags looking for real bugs.
>    

No, this is GCC being stupid.

> I would suggest we modify the coding style to ask people to not init a
> struct like this.
>    

How else do you terminate a list?  IOW:

MyDeviceInfo device_infos[] = {
   {"foo", 0, 2},
   {"bar", 0, 1},
   {} /* or { 0 } */
};

This is such a pervasive idiom that there's simply no way that GCC can 
possibly try to warn against this.  Plus, it's entirely reasonable.

I think this is just a false positive in GCC.  Otherwise, there's a ton 
of code that it should be throwing warnings against.

Regards,

Anthony Liguori

> Cheers,
> Jes
>